Microbead-Based Ligase Detection Reaction Assay Using a Molecular

Dec 11, 2013 - The packed bead sample was then scanned by a fluorescence scanning imager to detect the presence of any mutations. With the developed m...
1 downloads 10 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Microbead-Based Ligase Detection Reaction Assay Using a Molecular Beacon Probe for the Detection of Low-Abundance Point Mutations Sho Watanabe, Kenta Hagihara, Kazuhiko Tsukagoshi, and Masahiko Hashimoto* Department of Chemical Engineering and Materials Science, Faculty of Science and Engineering, Doshisha University, Kyotanabe, Kyoto 610-0321, Japan ABSTRACT: A microbead-based ligase detection reaction (LDR) assay using a molecular beacon probe was developed for the facile and rapid detection of point mutations present in low copy numbers in a mixed population of wild-type DNA. Biotin-tagged ligation products generated in the LDR were captured on the surface of streptavidin-modified magnetic beads for purification and concentration. The resulting producttethered microbeads were combined with a molecular beacon probe solution, and the suspension was directly flowed into a capillary. The microbeads were accumulated in a confined space within the capillary using a bar magnet. The packed bead sample was then scanned by a fluorescence scanning imager to detect the presence of any mutations. With the developed methodology, we were able to successfully detect one cancer mutation in a mixture of 400 wild-type templates (t test at 95% confidence level). Furthermore, the post-LDR processing, typically the most laborious and timeconsuming step in LDR-based mutation detection assays, could be carried out much more rapidly (approximately 20 min). This was enabled by the simple bead and fluid manipulations involved in the present assay.

H

because signals from rare mutations are masked by those from the wild-type sequence majority.7 One promising technique that could overcome the limitations of direct sequencing is the ligase detection reaction (LDR). This reaction can distinguish specific sequence variations even in the presence of a majority of nonvariant DNA. In the LDR assay, a successful ligation event between two primer ends (a common primer and a discriminating primer) can occur only if the primers are perfectly complementary to the target DNA, especially at the 3′-end of the discriminating primer. The use of a thermostable DNA ligase allows one to subsequently employ thermal cycling to amplify the ligation products. Selective detection of the LDR products to signal the presence of a mutation has conventionally been performed using either electrophoretic separation3,8,9 or low-density DNA microarrays.10,11 However, electrophoretic separation is laborintensive and time-consuming, as typical slab gel electrophoresis requires a minimum of several hours for gel preparation, electrophoretic sorting and gel imaging. Microarray-based analysis, which demands expertise in the field and the use of high-precision robotics for array preparation and posthybridization image acquisition, is even more laborious and time-consuming, as the hybridization reaction alone normally takes >12 h to run to completion.12 Recently, there has been a notable trend toward the use of microspheres or beads as a platform for surface binding-based

uman genome-wide mutation screening is still extremely demanding even when state-of-the-art technologies are employed. It therefore remains necessary to be able to target known mutations at specific loci. Many genetic diseases are associated with point mutations in particular genes; these include single-base substitutions, deletions, and insertions.1 For example, K-ras gene mutations have been observed in various cancers such as pancreatic cancer,2 colorectal cancer (CRC),3 lung cancer,4 and gastric cancer.5 In colorectal cancer, most Kras mutations are localized on codon 12, and to a lesser degree on codons 13 (exon 1) and 61 (exon 2), and occur early in the development of the disease in 30−50% of patients.3 Once acquired, these K-ras mutations remain conserved throughout tumor development and progression. Therefore, such mutations can be excellent diagnostic markers for clinical staging (prognosis), early detection, and prediction/monitoring of treatment courses. A major bottleneck in mutation detection is that the mutation of interest may be present in low copy numbers in a mixed population of higher copy number, wild-type DNA. Even at the primary tumor site for many cancers, the normal wild-type stromal cell content can be as high as 70%. Therefore, if the mutation is found on only one of the two chromosomes of a tumor cell (heterozygous), the mutant DNA content can be as low as 15%.6 This number decreases precipitously the further away from the primary tumor site the sampling is conducted. Direct sequencing is an effective tool that allows simultaneous identification of the location and allelic composition of a DNA sequence. However, conventional sequencing techniques are not suitable for cancer detection © 2013 American Chemical Society

Received: November 1, 2013 Accepted: December 11, 2013 Published: December 11, 2013 900

dx.doi.org/10.1021/ac403531x | Anal. Chem. 2014, 86, 900−906

Analytical Chemistry

Article

assays and analysis. Several reasons for the popularity of beadbased analysis are as follows: (1) Capture of the target analytes on the bead surface simplifies separation from the reagent/ nontargeted analytes in solution. For example, beads can be separated from the supernatant by a simple centrifugation step or through the use of an external magnet (for magnetic beads). (2) Because the effective ratios of reactive surface area to volume are large, it becomes possible to perform surface binding reactions in much smaller volumes compared with those used for planar binding substrates such as a glass or plastic slides, as are employed in conventional DNA microarray studies. (3) Increased concentrations of analyte bound on the bead surface relative to that in solution can result in higher signals compared with the same reaction in solution. Thus, beads can effectively produce an amplified signal. To date, many types of microsphere surface modifications have been designed for the localization of various biomolecules through the use of specific interactions. These developments have been accelerated by an increasing demand from researchers, who employ particle surfaces as a sensing platform for a wide range of applications. The current spectrum of biodiagnostic applications using microspheres is enormously broad and includes immunoassays,13 affinity assays,14 DNA hybridization assays,15 and studies of protein−protein,16 protein−peptide,17 and protein−DNA18 interactions. Here we describe a microbead-based LDR assay developed for the highly sensitive detection of low-abundance singlenucleotide mutations in human genomic DNA. The assay strategy for detecting the targeted mutant sequences is illustrated in Figure 1. In this assay, the ligation is performed in a tube using a biotinylated common primer and an allelespecific discriminating primer with a mixture of polymerase chain reaction (PCR) amplicons, which are derived from wildtype and mutated genomic DNA templates. The amplicons are mixed with two LDR primers that flank the mutation of interest. The discriminating primer contains a nucleotide at its 3′ end that corresponds with the single-base mutation site. If there is a mismatch (e.g., C/T mismatch, as shown in the outcome on the left in Figure 1), ligation of the two primers does not occur. However, a perfect match (e.g., A/T match as shown in the outcome on the right in Figure 1) results in successful ligation of the two primers, producing a ligation product (i.e., LDR product). Biotin-tagged products generated from successful ligations in the presence of the target mutation are then immobilized on streptavidin-conjugated magnetic beads through the biotin−streptavidin interaction. After the beads are separated from the supernatant through the application of a magnetic field, a molecular beacon (MB) probe, which is a single-stranded oligonucleotide with a stemloop structure that becomes fluorescent when bound to a complementary sequence,19 is mixed with the beads to monitor the LDR products isolated on the bead surfaces. The MB used in this study was labeled with a fluorophore and a nonfluorescent (dark) quencher at the 5′ and 3′ ends, respectively, and contained a sequence within the loop region that could form a duplex with a complementary sequence within the discriminating primer (i.e., the target sequence for the MB). In the absence of the target sequence on the bead surfaces (as shown in the outcome on the left in Figure 1), the MB does not fluoresce because the hairpin stem keeps the fluorophore close to the quencher. However, when the probe sequence in the loop anneals to the target (as shown in the outcome on the right in Figure 1), the rigidity of the probe-

Figure 1. Illustration of the microbead-based LDR assay. The wildtype and mutant template DNA is denatured at 94 °C, and the two LDR primers (the common primer (CP) and the discriminating primer (DP)) are annealed to the target strand at 60 °C. The two primers are covalently joined by a ligase enzyme to form a continuous oligonucleotide strand (i.e., the LDR product) but only if the primers are perfectly complementary to the target. The molecular beacon (MB) probe is designed such that it incorporates a sequence complementary to a region within the discriminating primer. Upon formation of the LDR product−probe duplex, which is thermodynamically more stable than the stem-loop conformation of the MB, the probe undergoes a conformational change that restores the probe fluorescence. This signals a successful ligation event (as shown in the outcome on the right). If the DNA template has even a one base-pair mismatch with the discriminating primer at the 3′ end, the product will not be ligated and no fluorescent signature will be produced (as shown in the outcome on the left).

target hybrid forces the hairpin stem to unwind, separating the fluorophore from the quencher, thereby restoring fluorescence. Using the above strategy, our bead-based LDR assay can be used to signal the presence of rare mutations in the presence of a majority of wild-type alleles. In addition, the assay presented here can be performed rapidly and easily because only simple bead and fluid manipulations are involved in the post-LDR processing steps. 901

dx.doi.org/10.1021/ac403531x | Anal. Chem. 2014, 86, 900−906

Analytical Chemistry

Article

Table 1. Oligonucleotide Sequences Used for PCR and LDR oligos

usage

sequences (5′ → 3′)

size (mer)

K-ras Forward K-ras Reverse K-ras c12 Com-2 K-ras discr. c12.2WtG K-ras discr. c12.2D K-ras discr. c12.2A K-ras discr. c12.2V molecular beacon (MB) Tmpl. Wt G12G Tmpl. Mut G12D Tmpl. Mut G12A Tmpl. Mut G12V

PCR PCR Cmn. primer Discr. primer Discr. primer Discr. primer Discr. primer Probe DNA LDR template LDR template LDR template LDR template

TTAAAAGGTACTGGTGGAGTATTTGATA AAAATGGTCAGAGAAACCTTTATCTGT paTGGCGTAGGCAAGAGTGCCT-Biotin AAACTTGTGGTAGTTGGAGCTGG AAACTTGTGGTAGTTGGAGCTGA AAACTTGTGGTAGTTGGAGCTGC AAACTTGTGGTAGTTGGAGCTGT b Cy5-CTCTTCTAGCTCCAACTACCACAAAGAAGAG−IAbRQc AGGCACTCTTGCCTACGCCACCAGCTCCAACTACCACAAGTTT AGGCACTCTTGCCTACGCCATCAGCTCCAACTACCACAAGTTT AGGCACTCTTGCCTACGCCAGCAGCTCCAACTACCACAAGTTT AGGCACTCTTGCCTACGCCAACAGCTCCAACTACCACAAGTTT

28 27 20 23 23 23 23 31 43 43 43 43

p, Phosphorylated. bCy5, λex = 648 nm, λem = 668 nm. cIAbRQ, Iowa Black RQ as a quencher molecule (λabs = 656 nm). The boldface sequence in the discriminating primers is complementary to the loop region sequence of the MB. The underlined sequence comprises the MB stem, while the remaining sequence corresponds to the loop region. The boldface letters in the template DNA sequences represent the target nucleotides that differ from the wild-type sequence. a

■

were thermally cycled 20 times for 30 s at 94 °C and 2 min at 60 °C using a commercial thermal cycler. The reaction was then terminated by rapid cooling to 4 °C. Purification and Concentration of the LDR Products Using Magnetic Beads. We used 2.8-μm diameter streptavidin-coated magnetic beads (Dynabeads M-270 Streptavidin, Life Technologies, Carlsbad, CA) as solid supports for isolating the biotin-tagged ligation products generated in the LDR step. Capture of the LDR products on the beads was performed using the biotin−streptavidin interaction, according to the following procedures. A total of 10 μL of 10 mg mL−1 (i.e., 6−7 × 108 microbeads mL−1) stock bead suspension was washed with 100 μL of 10× SSC buffer twice and then resuspended in 50 μL of 10× SSC buffer. A total of 50 μL of the LDR product solution was then combined with 50 μL of the washed bead suspension, followed by gentle mixing for 10 min at 25 °C by a rotator (Eyelaco, Tokyo, Japan). After immobilization, the beads were washed with 100 μL of 10× SSC buffer three times to remove unbound biotinylated oligonucleotides. Fluorescence Detection of the LDR Products on the Beads Using an MB Probe. An MB probe was labeled with Cy5 (reporter dye) and Iowa Black RQ (quencher) at the 5′ and 3′ ends, respectively; its sequence is described in Table 1. The product-tethered beads mentioned above were resuspended in a 0.2 mL microcentrifuge tube with 20 μL of 10× SSC buffer. For hybridization of the MB to its target sequence, 20 μL of 1 μM MB solution was added to 20 μL of the bead suspension, followed by gentle mixing for 1 min. The resulting bead suspension was fed into a fused-silica capillary (length, 20 cm; i.d., 100 μm; o.d., 200 μm; GL Sciences, Tokyo, Japan) by pressure-driven flow, followed by introduction of 5× SSC buffer. This sequential fluid operation enabled bead accumulation of a constant amount (∼ 90 μg or 6 × 106 microbeads) at the 1-mm-thick neodymium magnet bar (NeoMag, Chiba, Japan) placed on the side of the capillary (see Figure 2). Prior to bead loading, the outer-wall polyimide coating of the capillary was removed 10 cm away from the capillary end to serve as an ∼10-mm-long detection window for fluorescence detection. Packed bead samples in separate capillaries were aligned on a glass plate with a constant interspacing of 1 mm. To acquire fluorescence images of the packed beads, the regions around

EXPERIMENTAL SECTION PCR amplification of genomic DNA. PCR amplifications were carried out to generate 290 bp amplicons using 50 μL of 1× GoTaq reaction buffer (pH 8.5, 1.5 mM MgCl2; Promega, Tokyo, Japan), 300 μM dNTPs, 0.5 μM forward and reverse primers, and 50−200 ng of genomic DNA extracted from cultured human cell lines of known K-ras genotype associated with the onset of colorectal cancer (SW620 for mutant G12 V and HT29 for wild-type G12G; ATCC, Manassas, VA). The nomenclature of the given mutation (e.g., G12 V) denotes a DNA base substitution (G → T) within codon 12 (wild-type: GGT) in exon 1 of the K-ras gene, which alters the amino acid at this position from glycine (G) to valine (V). The genespecific PCR primers were obtained from Integrated DNA Technologies (IDT; Coralville, IA) and their sequences are described in Table 1. After a 2-min initial denaturation at 94 °C, 1.25 U of GoTaq DNA polymerase (Promega) was added under hot-start conditions and amplification was achieved by thermal cycling for 35 cycles at 94 °C for 15 s, 60 °C for 30 s, 72 °C for 18 s, and a final extension at 72 °C for 1 min, using a commercial thermal cycler (Eppendorf Japan, Tokyo, Japan). PCR products were quantified by absorbance at 260 nm and stored at −20 °C until required for the LDR. The sequences of the PCR amplicons adjacent to the target alleles were confirmed by an offsite DNA sequencing service (SigmaAldrich Japan, Tokyo, Japan). LDRs. LDRs were performed using conditions similar to those described elsewhere with slight modifications.9 In brief, a mixture of the aforementioned PCR amplicons (wild-type and mutant DNA) were used as template DNA for the LDR through addition to a 50-μL solution containing 1× Taq DNA ligase buffer (20 mM Tris-HCl, 25 mM potassium acetate, 10 mM magnesium acetate, 1.0 mM β-nicotinamide adenine dinucleotide (NAD+), 10 mM dithiothreitol, and 0.1% Triton X-100) at pH 7.6 (New England Biolabs, Beverly, MA), 100 nM biotinylated common primer (IDT), and 100 nM Cy5labeled discriminating primer (K-ras discr. c12.2WtG, c12.2D, c12.2A, or c12.2V; IDT). See Table 1 for the sequences of the common primer and the discriminating primers. The potassium ion concentration was adjusted to 100 mM through addition of an appropriate amount of potassium chloride. After an initial 2min denaturation step at 94 °C, 20 U of Taq DNA ligase (New England Biolabs) was added to the cocktail, and the reactions 902

dx.doi.org/10.1021/ac403531x | Anal. Chem. 2014, 86, 900−906

Analytical Chemistry

Article

Figure 2. Microscopic image of accumulated beads in the capillary. Prior to bead loading, the outer-wall polyimide coating of the capillary was removed to serve as a detection window for fluorescence detection. The inset shows an enlarged view of the packed beads.

the packed beads were scanned by a laser-induced fluorescence scanning imager (FLA-8000, Fujifilm, Tokyo, Japan) at an excitation wavelength of 635 nm.

■

RESULTS AND DISCUSSION Hybridization Specificity of the Designed MB Probe. The MB probe should be able to bind specifically to the target sequence, and the hybrid formed between the probe and the target should be more stable than intramolecular stem pairing. To demonstrate the hybridization specificity of the designed MB probe, we compared fluorescence intensities for the probe at ambient temperature in the presence of the target sequence (synthetic LDR product strand or the discriminating primer), a noncomplementary sequence (the common primer), or no target (i.e., MB only) using a spectrofluorometer (Figure 3A). The nonhybridized probe was very weakly fluorescent due to the hairpin structure bringing the quencher in close proximity to the reporter dye. In contrast, the probe became strongly fluorescent when mixed with the synthetic LDR product or the discriminating primer, indicating that the MB spontaneously bound to its target, dissociating the stem, thereby resulting in fluorescence. When the common primer was added to the MB solution, a weak fluorescence intensity was observed similar to that obtained with MB alone. Therefore, it can be concluded that the MB probe used in our study was able to specifically hybridize to the target sequence contained within the discriminating primer sequence. The significant difference in the fluorescence intensity between the positive and negative controls even at ambient temperature indicates that removal of unbound MB from the solution after incubation with the target-tethered beads is unnecessary. This is a clear advantage over conventional DNA− DNA hybridization techniques using a dye-tagged oligonucleotide probe, as this approach normally requires stringent posthybridization washes for the removal of nonspecifically bound probes as well as careful temperature control to obtain sufficient hybridization fidelity and efficiency. It should be noted that although unbound probes remaining in solution phase can be removed much quicker in a bead-based assay, this method cannot necessarily guarantee the complete removal of nonspecifically bound probes. The sufficiently low fluorescence intensity of the negative control in the above experiments is most likely a direct benefit of the use of the MB as a probe, as the MB appears to exhibit negligible binding to the nontargeted DNA (i.e., the common primer in this particular case; see the outcome on the left in Figure 1). Therefore, the use of MB is still an attractive advantage of our

Figure 3. (A) Comparison of fluorescence intensities for the MB probe in the presence of a target sequence (43-mer synthetic LDR product strand (Tmpl. Mut G12D) or the discriminating primer (Kras discr. c12.2D)), a noncomplementary sequence (the common primer), or no target (i.e., MB only). The sequence of the synthetic LDR product used was a combination of the discriminating primer and the common primer sequences (see Table 1 for all sequences). The fluorescence value of the sample that offered the highest counts was scaled to 100 for normalization. (B) Comparison of fluorescence intensities for packed bead samples with different hybridization times of 1, 10, and 40 min. The total fluorescence count at a confined area (100 μm × 500 μm) within the imaged bead sample was measured. The integrated fluorescence value for the sample with 1-min hybridization was scaled to 100 for normalization.

approach compared with conventional DNA−DNA hybridization analyses, even where microbead-based techniques are also employed in the latter to simplify the separation of unbound probes from bound ones in solution. Rate of Binding of the Designed MB Probe. The above binding specificity test was carried out in solution phase; however, the probe hybridizations carried out in this study occurred on bead surfaces. It is well-known in the DNA microarray field that the kinetics of complementary hybridization or sequence-specific binding of probes and target molecules are generally slower when one strand is covalently bound to a solid phase, because of restricted movement compared with the three-dimensional solution phase.20 We therefore investigated the optimal incubation time for adequate binding of the MB probe to the bead-captured target. First, LDRs were performed using a biotinylated common primer and a discriminating primer (K-ras discr. c12.2 V in Table 1) with a synthetic complement to the primers as the template DNA (Tmpl. Mut G12V in Table 1). After immobilization of the resulting LDR products on the bead surfaces, the bead suspensions were combined with MB solution and gently mixed for 1, 10, and 40 min using a rotator. Each set of beads was packed into separate capillaries 903

dx.doi.org/10.1021/ac403531x | Anal. Chem. 2014, 86, 900−906

Analytical Chemistry

Article

that were subsequently scanned using a fluorescence image analyzer to compare the fluorescence readouts. As shown in Figure 3B, similar fluorescence values were observed across all three samples, indicating that an incubation period of 1 min is sufficient for complete annealing of the MBs to the beadtethered target. This hybridization time is much shorter than that employed in conventional DNA microarray studies (∼12 h), which could be explained by increased probe molecule transport to the bead surface compared with stationary, planar surfaces: beads can be moved simply through agitation. However, even in the absence of such movement, beads are still not completely stationary because of their Brownian motion resulting from the random thermal motion of the surrounding solvent molecules. Hence, the solution at the bead surface is continuously being refreshed and the whole binding process occurs under conditions much closer to those in a free solution, improving the binding kinetics. This in turn reduces the processing time of the bead-based assay. Assay Specificity. Beads tethered with the correct ligation products generated on matched templates need to exhibit significantly higher fluorescence signals over the background noise generated from misligation on mismatched templates, so as to identify the nucleotide at the target position with a high degree of certainty. To examine the signaling fidelity of the present assay, we prepared four synthetic template DNAs mimicking exon 1 mutations in the K-ras gene that differed from one another at one nucleotide position only (Tmpl. Wt G12G, Mut G12D, G12A, and G12 V; see Table 1). For each template DNA, four LDR cocktails were prepared, each containing one of the four discriminating primers, each differing at one nucleotide position at the 3′ end (K-ras discr. c12.2WtG, c12.2D, c12.2A, or c12.2V; see Table 1). Therefore, it was anticipated that for each LDR set, ligation would be initiated with only the one discriminating primer that was perfectly complementary to the template DNA present. Figure 4 shows the acquired fluorescence images of packed beads derived from LDR samples using all possible combinations of the four templates and the four discriminating primers (Figure 4A) and their lane integration results (Figure 4B). In lane (v) for the blank test (i.e., no template), signal counts were slightly raised at each bead position, which was probably due to light scattering by the packed beads, as any unbound MB was flushed out during the bead packing step. Similarly low signals were also observed at three of the bead positions in each of the four test lanes containing template, indicating that very low amounts of mismatched ligation products were generated. The first bead position in lane (ii) gave a slightly higher signal than that of the other mismatched samples, which is attributable to the ligase enzyme discriminating the T/G mismatch with lower fidelity than other mismatches.21 However, the fluorescence signals derived from the matched ligation products were all at least 15 times higher than that of the T/G mismatch, and thus high signaling fidelity of the present assay was confirmed. The results of this specificity test also show that a single MB probe can be used to identify all possible allelic variations at a target nucleotide. This is because the MB probe is designed to hybridize to the common sequence contained within the ligation products (i.e., the boldface sequence in the discriminating primers in Table 1). Detection of a K-ras Mutation in a Mixture of Majority Wild-Type DNA. We next developed an assay to demonstrate the ability of our platform to detect mutant DNA in the

Figure 4. Assay fidelity test using different combinations of template DNA and discriminating primer for LDRs. (A) Fluorescence images of beads packed in separate capillaries. Synthetic template DNA used for preparing the samples depicted in lanes (i), (ii), (iii), and (iv) were Tmpl. Wt G12, Mut G12D, Mut G12A, and Mut G12 V (Table 1), respectively. No template was used in the negative control shown in lane (v). The discriminating primer incorporated at each respective bead position is shown along the horizontal axis. (B) Integration of fluorescence signals over lanes (i)−(v).

presence of an excess of wild-type DNA. To realize a situation close to clinical diagnosis, normal DNA and mutant DNA derived from cell lines of known K-ras genotype (wild-type G12G and mutant G12 V) were PCR amplified independently and then mixed, allowing a range of mutant to wild-type ratios to be tested. Figure 5A displays an acquired image for bead samples prepared using LDR cocktails containing various dilutions of mutant DNA into wild-type DNA, while Figure 5B shows integrated fluorescence counts for the imaged bead regions. When the discriminating primer for the mutant DNA was applied to the mismatched wild-type template alone as a negative control, a significantly higher signal than that for the blank sample (no template) was observed. This signal was also at a similar level to that obtained for the matched samples in the fidelity test in Figure 4. The increased background signal count in this experiment is attributable to the greater number of misligation products (C/T mismatches) generated because of the 100-fold higher concentration of mismatched template used (10 nM) compared with that employed in the fidelity test (0.1 nM). When testing the same discriminating primer on a mixture of mutant and wild-type PCR amplicons at a mutant to wild-type ratio of 1:10, a fluorescence signal representing matched (A/T) ligation products was obtained that was significantly larger than 904

dx.doi.org/10.1021/ac403531x | Anal. Chem. 2014, 86, 900−906

Analytical Chemistry

Article

electrophoresis and DNA microarrays, are laborious and timeconsuming. The best previously reported microarray-based LDR assay, which was assisted by microfluidic technology, could complete the post-LDR procedures such as hybridization, washing, and scanning, in only 20 min.11 The reduced processing time was mainly attributed to the improved hybridization kinetics from constructing the probe array in the microfluidic channel. However, upstream array preparation took at least several hours, limiting the practicality of this approach. In contrast, the bead-based LDR assay developed here involves only simple bead and fluid manipulations after the completion of the LDR phase. The use of an MB probe enables direct detection of the LDR products tethered to the bead surfaces. Because of these attributes, the post-LDR processing could be carried out rapidly: 10 min for immobilization of the LDR products onto the beads, 3 min for bead washing, 1 min for MB probe hybridization, 5.5 min for the bead packing, and 1 min for scanning (∼20.5 min total). Using the developed methodology, we could successfully detect one cancer mutation in a mixture of 400 wild-type templates. In addition, this platform is highly amenable to multiplexing, which would enable the simultaneous monitoring of multiple mutations of interest.

Figure 5. Detection of point mutations using different mutant to wildtype template ratios. (A) Images of packed beads in capillaries obtained with a fluorescence scanner. The numerical annotations represent the mutant to wild-type ratios. (B) Comparison of total fluorescence counts for the scanned bead areas. The experimental conditions were the same as in Figure 4, except that PCR amplicons were used as template DNA for the LDRs at a fixed wild-type concentration of 10 nM, instead of synthetic oligonucleotides at a concentration of 0.1 nM. The error bars represent standard error of the mean, n = 3. The numerical annotations above the bars represent the calculated t values to test for differences between the positive and negative control samples. *Significant at p < 0.05; n.s., not significant.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +81 774 65 6594. Fax: +81 774 65 6803. Notes

The authors declare no competing financial interest.

■

that of the negative control (t test p-value